Analysis of DAWNBench, a Time-to-Accuracy Machine Learning Performance Benchmark

Operating Systems Review, Volume abs/1806.01427, Issue 1, 2019.

Cited by: 46|Views163
EI

Abstract:

The deep learning community has proposed optimizations spanning hardware, software, and learning theory to improve the computational performance of deep learning workloads. While some of these optimizations perform the same operations faster (e.g., switching from a NVIDIA K80 to P100), many modify the semantics of the training procedure (...More

Code:

Data:

Your rating :
0

 

Tags
Comments